Survival Disparities By Insurance Type For Patients With Non-Hodgkin Lymphoma

Lack of access to health insurance is known to be a poor prognostic indicator in many conditions. Here, we examine survival in patients with non-Hodgkin lymphoma (NHL) in the US by insurance status, including no insurance, Medicaid only, and other insurance (private or Medicare) to explore the extent of this effect in NHL patients.

Data were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Information regarding patients' insurance status became available in the SEER database in the most recent dataset and is available from 2007 on. The SEER17 database was used to provide the maximum number of patients. Period analysis was performed to estimate 1-year survival and complete analysis was used to estimate 3-year survival. Patients with a diagnosis of NHL diagnosed between 2007 and 2010 with follow up available through 2010 were included. Because patients over age 65 are almost all qualified for Medicare, patients age 15-64 were evaluated. Because NHL is a heterogeneous condition with subtypes having varied survival, we examined survival in diffuse large B-cell lymphoma (DLBCL) specifically as well.

Relatively few patients were listed as uninsured, with less than 10% of patients being listed as uninsured in the SEER database (as opposed to over 20% in the US population at large for patients under age 65.) Initial age stratification showed that there was very little difference in survival for patients age 45-64 and therefore these patients were considered as a single age category to improve statistical power.

For NHL overall, survival was much lower for both uninsured patients and patients with Medicaid as compared to patients with other forms of insurance at both 1- and 3-years and for each age group examined (Table). For patients age 15-44, 1-year relative survival was 15.1 and 17.7 percentage points lower for uninsured and Medicaid patients, respectively, than for patients with other forms of insurance. The differences for 3-year survival were 18.8 and 22.9% units, respectively, uninsured and Medicaid patients. Three year relative survival for patients age 15-44 without insurance was 67.8%. For comparison, 3-year survival for the same age group in 1978-80 was 69%.

A similar pattern was seen for patients with DLBCL, with survival being higher for patients with private insurance compared with patients who were uninsured or insured by Medicaid. One year survival estimates were 12.8 and 13.4% units lower, respectively, for uninsured patients ages 15-44 and 45-64 as compared to patients with private insurance or Medicare. Three year survival estimates were 14.2 and 14.1% units higher, respectively, for the same comparison.

Lack of insurance was associated with severe compromise of survival for patients with lymphoma. Survival is much lower for patients with Medicaid as opposed to other forms of insurance, possibly because of the extreme poverty required for Medicaid eligibility in most states, which may compromise patients' ability to be compliant with care. Furthermore, patients with Medicaid may have only been granted Medicaid after the diagnosis of cancer and been previously uninsured, which may be a risk for presentation at a later stage of disease. Further evaluation of the reasons for the low survival for patients with Medicaid and implementation of comprehensive coverage for medical care for uninsured patients are urgently needed to reduce this disparity.

No relevant conflicts of interest to declare.